Loading ToolBench

Reading the current corpus…

f1-mcp MCP server · C · ToolBench

← Browse

f1-mcp

github.com/TommyKisly/f1-mcp

Live Formula 1 data over MCP: championship standings, race results, and the season calendar, from the jolpica-f1 API.

2025-06-18+STREAMABLE_HTTP3 tools

65/100

Definition Quality83

Protocol Readiness81

Supportability25

Spec posture

Inferred effective spec 2025-06-18+·current baseline 2026-07-28

Static source inference · medium confidence · evidence: Apps extension, structured output, Streamable HTTP

No deprecated protocol patterns detected

Summary

Three tools with clear verb-noun naming (get-*), good descriptions (120-180 chars), and complete input/output schemas. Tool annotations present (readOnlyHint, destructiveHint). However, parameter descriptions are minimal or absent in some cases, and error handling lacks recovery guidance. Output schemas are well-typed but responses could be more concise. Composition is sound, each tool has one responsibility and outputs chain naturally.

Tools (3)

get-driver-standingsread onlysource verified85/100

Current Formula 1 drivers championship standings: position, points, wins, and team for every driver this season. Renders a standings board widget.

get-race-resultsread onlysource verified82/100

Results of a Formula 1 race this season. Defaults to the most recent race. Pass a round number for an earlier one.

get-race-scheduleread onlysource verified82/100

The full current Formula 1 season calendar with dates, plus which race is next from today.

Issues (5)

MEDIUMconstrained-input ↗

get-race-results 'round' parameter lacks actionable format guidance. Description says 'Round number' but does not specify valid range, format (string vs int), or how to discover valid rounds.

get-race-results

MEDIUMrecovery-guide ↗

No error recovery guidance. If ergast() fails (e.g. 'no results for round yet'), the error message is returned as-is without suggesting next steps (e.g. 'Try get-race-schedule to find completed rounds').

get-race-resultsget-driver-standingsget-race-schedule

LOWresponse-shaper ↗

Output responses include raw API fields (e.g. 'grid', 'status' in race results) that may not be immediately useful to LLMs. Consider stripping verbose metadata or adding field descriptions to the output schema.

get-race-results

LOWpaginated-result ↗

No pagination or result limits documented. If a season has 24 races, all are returned in one response. For future scalability, consider adding limit/offset parameters and documenting max result size.

Recommendations

Add explicit format and range constraints to 'round' parameter: 'Round number (1-24, or omit for latest). Use get-race-schedule to discover available rounds.'
Enhance error messages with recovery paths: catch ergast errors and return 'No results for round X yet. Call get-race-schedule to see completed races.' instead of raw HTTP status.
Document output schema fields inline (e.g. 'grid: starting grid position (1-20)') so LLMs understand what each field represents.
Add per-tool timeout and rate-limit guidance in descriptions to prevent agent runaway loops.
Consider adding a 'get-driver-details' tool to enable multi-step queries (e.g. standings → select driver → fetch career stats) without requiring external lookups.

Spec posture evidence

Inferred effective spec: 2025-06-18+.

Current-spec patterns present: structured tool output, tool annotations, Streamable HTTP

Score history

Overall score trend

First recorded score · v2 rubric

65/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	C	65	2025-06-18+	v2